INDEX

Explanations

references to government spending and legislative secrecy

oai_token-act-pair · gpt-4o-mini Triggered by @bot

specific text and formatting

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

routeProvider

-0.32

 confront

-0.31

 confronting

-0.30

is

-0.30

unt

-0.29

主

-0.29

</th>

-0.28

 spor

-0.28

trans

-0.28

 confronted

-0.28

POSITIVE LOGITS

UserScript

0.75

rungsseite

0.65

CodedInputStream

0.63

majánló

0.61

antası

0.60

 ſtate

0.59

 ErrIntOverflow

0.57

 secretly

0.57

ロウィン

0.56

 laſt

0.55

Activations Density 0.028%

references to government spending and legislative secrecy

specific text and formatting

No Comments

No Known Activations